A 3D FINITE ELEMENT MODEL OF THE FEMALE PELVIC FLOOR FOR THE RECONSTRUCTION OF URINARY ...congress.cimne.com/iacm-eccomas2014/admin/files/file... · 2014-03-14 · Key words: Urinary

11th World Congress on Computational Mechanics (WCCM XI)5th European Conference on Computational Mechanics (ECCM V)

6th European Conference on Computational Fluid Dynamics (ECFD VI)E. Onate, J. Oliver and A. Huerta (Eds)

A 3D FINITE ELEMENT MODEL OF THE FEMALEPELVIC FLOOR FOR THE RECONSTRUCTION OF

URINARY INCONTINENCE

A. BHATTARAI∗, R. FROTSCHER∗, M. C. SORA† AND M. STAAT∗

∗ Biomechanics Laboratory, Institute of BioengineeringAachen University of Applied Sciences

Campus Julich, Heinrich-Mußmann-Straße 1, 52428, Julich, Germanye-mail: {bhattarai, frotscher, m.staat}@fh-aachen.de,

web page: http://www.fh-aachen.de/forschung/institut-fur-bioengineering

†Plastination Laboratory, Institute of AnatomyMedical University of Vienna

Wahringerstr. 13, A-1090, Vienna, Austriae-mail: [email protected],

web page: http://www.meduniwien.ac.at/sysanat/plast.html

Key words: Urinary Incontinence, Integral Theory, Pelvic Floor Anatomy, UrethralOpening, Active Closure.

Abstract. Urinary incontinence is a well known pelvic floor (PF) dysfunction in theaging female. Causes can be mainly assumed to be vaginal delivery and reduced stiffnessof supporting structures. These can be addressed effectively through a minimally invasivesurgery with an insertion of a Polyvinyl anchoring mesh ‘to restore the structure to im-prove its function’: the fundamental principle of the ‘Integral Theory System’. Althoughsignificant research has been done, an exact cause of incontinence remains elusive, eitherbecause of limited knowledge of pelvic floor anatomy or of a lack of anatomically realisticfinite element simulations. In order to cope up with such shortcomings and to facilitatean effective treatment method, the detailed study of pelvic anatomy, realistic modelingof pelvic floor functions and 3D finite element simulations of the pelvic floor dynamicsduring urethral closure and opening are the core interests of this research work.

1 INTRODUCTION

The anatomic structure of human pelvic diaphragm is considered to support pelvicviscera (organs) and to allow the passage of urine and feces. It is also responsible formaintaining the intra-abdominal pressure during the motions associated with daily phys-ical activities. Unlike the male, the female pelvic floor also sustains the pregnancy and

1


facilitates the fetal descent during delivery. From conception to birth, a mother’s bodyundergoes unique physical and mental changes. Current research on old female subjectshas dug out some evidences which can be traced from the injuries and severe damagesthat might occur during delivery.

It is well known that progesterone levels are extraordinarily high during delivery whichcauses a laxity of ligaments and muscles throughout the female body. This facilitates towiden the way out for fetal descend which normally regains its shape and functions aftersome time. However, critical damages are still vulnerable for next pregnancy or normalpelvic functions. Among them, pelvic organ prolapse and urinary incontinence (UI) arethe most common pelvic dysfunctions among the aging females greatly affecting theirquality of life. To the present day, only stress UI is said to be surgically curable whichis caused by loss of urethral support due to the damage of supporting structures (fascias,ligaments and muscles).

Following the contention of the Integral theory of Petros, the aim of this paper is toestablish a finite element model of the female pelvic floor based on patient data andvalidate it by FE simulations of the normal behavior of the pelvic floor. In this paper,emphasis is given on the detailed description of urethral dynamics and the structuresresponsible for it.

2 MATERIALS AND METHOD

The model was reconstructed from a 70 year old female cadaver specimen obtainedfrom a coroner. The specimen was a part of the human donation program of the MedicalUniversity of Vienna. Ethical committee of the university provided approval for using thepelvis for 3D reconstruction: EK Nr: 1191/2011.

2.1 Volume Rendered Computer Model

One female unfixed human cadaver pelvis was used for this study. After removal,the female pelvis was frozen at -80 ◦C for one week. The frozen tissue block was cutinto 1.5 mm thick slices. The slices were dehydrated for 3 weeks in cold acetone. Whendehydration was finished, the slices were degreased with methylenchlorid for 10 days. Thedehydrated/degreased specimens were submerged in an E12 impregnation mixture [1] ina Heraeus VT 6130 M vacuum chamber and impregnated for 2 days. After impregnation,the slices were removed out of the vacuum chamber and cast between glass plates. Theslices were then placed into an oven and hardened for the next 4 days at 45 ◦C.

In an additional step, we disassembled the glass plates and obtained the finished E12(epoxy) slices. Each plastinated slice was scanned from both sides using an EPSON GT-10000+ Color Image Scanner at 600 dpi. A ruler (mm) was included in every scan asa calibration marker. Scanned images of the plastinated tissue slices were imported asBMP files into the PC. These images were loaded into WinSURF soft1 for the reconstruc-

1SURFdriver 4.0; http:// www.surfdriver.com

2


(a) Ventral-superior view (b) Lithotomy position, caudal (inferior) view

Figure 1: A computer model of the female pelvic floor including bony pelvis, pelvic viscera, muscles,ligaments and nerves. (Reconstructed from plastinate by Medical University of Vienna).

tion step. The objects which had to be reconstructed, were traced manually by using agraphic table (Wacom Cintiq 24HD). Each object was traced and numbered accordinglyon every BMP file. After that, the reconstruction was rendered, visualized, and quali-tatively checked for surface discontinuities by rotating the model. Reconstructed objectwere transformed in .OBJ format as shown in Figure 1. Table1 lists the abbreviationsused in the figures.

Table 1: Abbreviations used in the figures

Abbreviation Full term Abbreviation Full termUSL Uterosacral ligament CL Cardinal ligamentPUL Pubourethral ligament PVL Pubovesical ligamentB Bladder U UrethraV Vagina R RectumUt Uterus Ur UreterN Nerves LA Levator aniPCM Pubococcygeus muscle ICM Illiococcygeus musclePRM Puborectalis muscle CCM Coccygeus muscleEAS External anal sphincter Cx CoccyxS Sacrum PS Pubic symphysisI Ilium

2.2 Geometrical Model

The next step is the mesh generation from the computer model. The model serves itsbest for the visualization of the pelvic floor anatomy. Anyhow, due to the complex geome-

3


try the reconstructed surfaces show intersecting faces, non-manifod edges and high aspectratio that makes a FE simulation impossible or at least affect the convergence. Smooth-ing and repairing the defects is done by using the 3D mesh processing software Meshlab2

and 3D analysis software Amira3. The repaired model is then imported into the opensource pre- and post-processor SALOME to create a finite element mesh. Consideringthe significant thickness of each pelvic constituents, we adopt a volume discretization. Asmooth FE mesh as shown in Figure 2 is constructed from linear tetrahedrons (>370,000elements) except for the vagina that has been discretized with quadratic tetrahedral ele-ments (>22,000) which are required for high precision due to contact between the anteriorand posterior walls and between the vagina and adjacent structures (rectum and bladder).The bony pelvis can be considered to be rigid thus it is excluded from the mesh in orderto reduce the computation time and cost.

(a) Lateral-ventral view in sitting position

PVL

U

V

PCM

PRM

ICM

R

EAS

CL

CCM

USL

B

PUL

(b) Inferior view in supine position

Figure 2: A 3D finite element mesh of the female pelvic floor based on the data set from the plastinatereconstruction.

Boundary ConditionsKinematic boundary conditions for the attachment of the pelvic constituents to the

bones are specified in terms of appropriate suppression of nodal displacements. Thedistal end of vagina and urethra are fixed to confirm the zero-displacement in every caseof pelvic function. Urethra and bladder are considered to be two different structuressharing common faces hence they are linked together. They are attached to pubourethraland pubovesical ligaments, which in turn are fixed at their other ends assuming the pubisinsertion. The lateral walls of the vagina stretch under the action of LA contraction whichare represented by directional forces4 [2]. The upper vagina is suspended at the level of

2http://meshlab.sourceforge.net/3http://www.amira.com4Currently, the pelvic muscles are not included in this work, which is replaced simply by applying

directional forces in the specific regions of the lateral wall of vagina. Based on the available geometricalrepresentation of the muscle, the muscle model will be included later in the simulation to transmit theforces during contraction and relaxation of LA muscle.

4


the cervical ring by the cardinal and uterosacral ligaments, the other ends of which arefixed to the pelvic side wall and sacrum, respectively.

3 THE ANATOMY OF PELVIC FLOOR

3.1 Pelvic Ligaments

The normal function of the female pelvic organs and maintenance of continence dependson the integrated system of (bony) pelvis, fibromuscular connective tissues, fibro-elasticligaments and muscular pelvic diaphragm [2], [3]. The pelvis comprises of bony scaffoldingconnected by a pair of coxal bones with sacrum and coccyx posteriorly and to each otheranteriorly at the pubic symphysis. The pubourethral ligament (PUL), a fan like structure,originates from the lower part of each pubic symphysis which bifurcates to insert intothe midurehra and lateral vaginal walls [4]. Based on the cadaveric dissection, physicalexamination in 30 patients presented some evidence of PUL together with muscle instabilizing urethra in its normal position and maintaining continence [5]. Some researchalso depicts these ligaments as extensions of the perineal membrane, as well as the caudaland ventral portions of the arcus tendineous fascia pelvis (ATFP) because of their intimateconnection. In fact, during sudden increase in intra-abdominal pressure, PUL tightlyanchors the anterior wall of the urethra thereby kinking the midurethra.

In [6] and [7] the authors refer to the pubovesical ligament (PVL) and the pubourethralligament to be synonyms for the same structure that is responsible for the vesical (bladder)neck functioning, urethral support and maintaining continence. Later in [8], histologicalexamination of the vesical neck region has been performed in order to differentiate theexistence of both ligaments in terms of orientation and functions: urethral support, conti-nence and initiation of micturition. Therein the existence of the PVL as bands of smoothmuscle fibres has been confirmed. It originates from the detrusor muscle at the vesicalneck which inserts into the anterior portion of the ATFP and pubic bone. It assist invesical neck opening and provide rigidity to the bladder [9].

The cardinal and the uterosacral ligaments suspend the apex of the vagina at the levelof the cervical ring and support the uterus. The cardinal ligament which is composedof smooth muscle attaches to the lateral pelvic wall to the obturator internus musclevia the obturator fascia. The USL contains a considerable amount of fibrous tissue andnon-striped muscular fibers that are attached to the sacrum posteriorly. Weakness andstretching of these ligaments can contribute to uterine prolapse: the drop of the uterusinto the vaginal canal.

3.2 Pelvic Diaphragm

Depending on the functional anatomy, skeletal muscles situated at the lower abdomencan be classified into two groups: (1) the obturator internus and the piriformis, whichform the pelvic side wall; (2) the coccygeus and levator ani, which together form thepelvic diaphragm [10]. The piriformis and the obturator internus muscles are responsible

5


(a) Superior view (b) Lateral view

Figure 3: FE mesh of the striated female pelvic floor diaphragm.

for hip flexion-extension during leg movement and play no role for the continence andstructural integrity of the pelvic viscera.

The coccygeus muscle as shown in Figure 3 supports the pelvic viscera. It originatesfrom the ischium spine at the sacrospinous ligament and inserts to the sides of the lowerpart of the sacrum and the upper part of the coccyx. The Coccygei also supports thecoccyx by pulling forward to control defecation. The levator ani is the important muscleof the pelvic diaphragm. The levator ani supports the abdominal pressure and stabilizespelvic organs by regulating the size of the levator hiatus. It plays a crucial role in thepreservation of the urinary and bowel continence. It comprises of three different muscles:

• Pubococcygeus Muscle: It is the fibromuscular layer of pelvic diaphragm whichoriginates approximately 1.5 cm above the inferior pubic symphysis and runs hor-izontally to insert into the distal vagina [6]. Fibres of PCM from each side sweepbehind the rectum to find its attachment at coccyx and sacrum. The strong pub-ococcygeus muscle contracts to create a forward force which hammock to close theproximal urethra and to control urine flow.

• Illiococcygeus Muscle: At the level of the PCM, illiococcygeal fibres complete thesuperficial layer of the pelvic diaphragm. It arises from the ischial spine and getsattached anteriolaterally at the junction of the arcus tendineus pelvic fascia andthe fascia of the obturator internus muscle. The ICM is supposed to provide majorsupport to the posterior compartment [11].

• Puborectalis Muscle: The PRM is the medial part of the levator ani which closes thepelvic cavity. The fibres of the PRM are oriented almost vertically from the lowerpart of the pubic symphysis and the superior fascia of the urogenital diaphragmon both sides and loop around the rectum like a sling at the level of the anorectaljunction [2]. The inferior fibres of the PRM around the rectum and the fibres ofthe longitudinal muscle of anus are anchored at the EAS muscle. PRM contracts

6


and pulls the levator plate upwards during sudden increase in abdominal pressurecreating a ventral bend between rectum and anal canal to prevent incontinence [12].

3.3 Pelvic Viscera

The female pelvic floor can be divided into three distinct spaces: (1) Urethrovesicalspace; (2) Vaginal space; and (3) Posterior compartment or rectal space. As shown in

Figure 4: FE mesh of the female pelvic floor without bony pelvis and muscles.

Figure 4, the urethra is directed obliquely downward and forward which connects theurinary bladder to the urethral orifice between clitoris and vagina. The lower two-third ofthe posterior urethral wall is connected to the anterior vaginal wall via the pubocervicalfascia (PCF). The anterior surface is firmly anchored to the symphysis pubis by thepubourethral ligaments and urogenital diaphragm. As the bladder has no inherent shape,a fully distended female bladder rests on the upper part of anterior vagina just belowthe cervix and creates an additional pressure during sudden increase in intra-abdominalpressure. Weakening or tearing of the PCF may cause cystocele. Behind the vagina therectum is located which is separated by the rectovaginal fascia.

4 URETHRAL CLOSURE AND OPENING: A COMBINED ACTION OFLIGAMENTS AND MUSCLES

The normal states of the urethra can be defined as: opening during micturition andactive closure (closure during effort) [2]. The dynamics of urethral closure and openingmechanisms have been described by Petros as a result of the contraction and relaxation ofPCM [2]. Three different perspectives were put forward: urethral, vaginal and mechanicalperspective. The main idea was that the relative position of the midurethra due to thedevelopment of muscle forces will distinguish the situation of active closure and opening.

During effort, due to a sudden increase in intra-abdominal pressure, LMA and PRMpull the levator ani muscle downward and backward. The displaced LA muscle pulls thevaginal and rectal attachment in the direction of deformation. This will allow the bladderto fall onto the vagina. In the same time, PCM contracts and provides a hammock to

7


the lateral vagina creating a forward force to push the midurethra. With an adequatelytight PUL the backward and downward forces are counteracted and the urethra is closedby kinking.

During urethral opening, the PCM relaxes against the PUL. It allows the whole effortsystem to drop back by LMA and PRM. Due to the conscious contraction of the detrusormuscle, the urethra relaxes to expel out the urine. If either one of the PUL and the PCMis weak by any reasons (e.g. damaged during delivery or reduced stiffness in old age), theposition of the urethra can not be stabilized at closure position. This might favors theurine to leak during sudden physical activities or in normal conditions sometimes. Thissituation is commonly called as urinary incontinence.

5 BIOMECHANICAL STUDY

Almost every soft biological tissues undergoes large deformation. Presence of abundantwater requires to consider incompressibility. Therefore, finite strain theories that considerincompressibility and hyperelasticity are used to describe the mechanical behavior. Inorder to reduce the number of parameters and to simplify the computation for a complexmodel like the pelvic floor, we utilized a widely used non-linear stress-strain relationship,the Mooney-Rivlin material model that is characterized by the strain energy function

W = C10(I1 − 3) + C01(I2 − 3), (1)

where C10 and C01 are the material parameters with dimensions of stress at lower andhigher level of strains and I1, I2 are the principal invariants of the right Cauchy-Greenstrain tensor. Lacking an alternative, the essential material parameters for each pelvicconstituents are adopted from different related literatures. Table 2 lists the materialparameters used for the following simulations.

Table 2: Biomechanical properties of pelvic ligaments and pelvic viscera. The mean C10 and C01 values(± SD) in MPa are given below

Type of Structure C10 C01 ReferenceRight uterosacral ligament 1.39 (± 1.4) 12.57 (± 13.57) [13]Left uterosacral ligament 1.73 (± 2.12) 8.61 (± 8.08) [13]Right broad ligament 1.03 (± 0.9) 8.84 (± 13.02) [13]Left broad ligament 0.62 (± 0.45) 4.53 (± 3.44) [13]Vagina 0.52 3.25 [14]Bladder 0.3 0.5 [14]

In the considered simulation, the mechanical properties of the cardinal ligament isdefined with regard to the study from [15], which is assumed to be equivalent to that of

8


the broad ligament. In this context the corresponding values for cardinal ligaments wereobtained from uniaxial tension tests performed in [13]. Lacking experimental data, thematerial parameters of USL are used for the PVL and the PUL as well. The comparisonof the numerical values prevails ligaments to be stiffer than vaginal tissues. Contrary tothat, data from [16] suggests that the ligament stiffness is much lower than that of thevagina. The variation might be due to multiple reasons: age and health of the subject,cadaver storage, ex vivo measurement conditions or experimental setup.

Under physiological conditions, the intra-abdominal pressure (IAP) is imitated by auniformly distributed load on the upper surface of the vesica. Previous research [17] ona similar pelvic model calculated an average IAP (9.65 kPa) for the subjects at standingpositions.

Depending on the functional anatomy as described in sections 2.2 and 3.2, the vagi-nal surface is acted upon by a set of horizontal and vertical forces to mimic the realisticphysiology of the pelvic floor. Petros [2] suggested that bladder neck closure is of primaryimportance in the maintenance of continence, hence our simulation is mainly centered onthe deformed states of urethra, bladder and vagina. For that purpose, a set of horizontaland vertical pelvic muscle forces are directly transferred to the vaginal walls which simpli-fied the model without pelvic diaphragm and rectum, see Figure 4. This not only reducethe large number of degrees of freedom but also lower the computation time becausecontact is not necessary to be computed between vagina and rectum.

6 RESULTS

The finite deformation simulations for the simplified mesh as shown in Figure 4 areperformed in the open source FE software, Code Aster5. The rectum is omitted becauseit has little influence on the simulated load cases. Loading conditions are applied inaccordance to the integral theory in order to achieve similar displacement of the structuresand to investigate the qualitative behavior of the model. For the material parameters thatgovern the simulation, we used the values given in Table 2. The simulation is performedin two distinct steps. In the first step, a combined action of the IAP, the LMA and thePRM are applied. Figure 5a shows the bladder and the vagina being displaced backwardand downward. The region of interest is magnified and rotated towards the anteriorface. It shows that certain stress is developed on the PUL due to the pull from the tightand descended vagina. As the proximal vagina is not directly attached to the urethra,significant tension would have been developed if the vaginal insertion of the PUL wasincluded in the model.

We observe stress concentration at the point of attachment of the ligaments to theorgans, especially in the case of the USL and the CL. Those stress concentrations re-sult from the fact that the attachments are modeled using boundary condition. Currentlythose stress concentrations seem to influence neither the accuracy of the result nor the con-

5www.code-aster.org

9


USL

CL

V

U

B

PVL

PULR. PUL

V

L. PUL

(a) Vertical descent of pelvic viscera (b) Contraction of PCM

Figure 5: Finite element simulation of the female pelvic floor showing regions of stress concentration.R. PUL denotes right PUL and L. PUL denotes left PUL.

vergence behavior. Nevertheless, in future simulations those boundary conditions mightbe replaced in order to improve the model. Contact between the vaginal walls has beensuccessfully simulated though is not visible from the given figure.

In the next step, the PCM contraction is activated by applying a forward pull onthe lateral wall of vagina. Forces associated with the first part of the simulation are keptconstant. Doing so, the urethra is pushed against the PUL. High stresses are at the lateralwall of the vagina due to the applied PCM force, see Figure 5b. Moreover, the model isunder current development and improvements in the additional structures, boundary andcontact conditions are still to reduce numerical issues in order to be able to apply largerand more physiological forces.

7 CONCLUSIONS

The focus of this contribution is mainly to investigate the performance of the threedimensional computational model of the female pelvic floor, to model the mechanicalprocesses and to improve the understanding of normal urinary dynamics as suggested byPetros [2]. The model is developed from a realistic anatomic geometry from plastinatereconstruction. All possible pelvic constituents and an appropriate muscular frameworkclose to the data set produced by Janda in [18] are included in the model. Simulationresults show the importance of the pubococcygeus muscle in the maintenance of normalurinary continence, moreover, a strong anchorage of the pubourethral ligament is requiredindeed. The presented model will be further developed in order to simulate functionaldefects in the pelvic floor and to investigate the quality of respective mesh implantswith respect to their ability of recovering incontinence. In related projects we alreadyinvestigated different meshes and anchoring systems experimentally and computationally[19], [20]. The structure and the material of different meshes varies considerably andtherefore the question arises which meshes are preferable to use. One purpose of themodel is to develop criteria or quality measures for the meshes via simulations. Moreover

10


the model can be used for the optimization of the mesh implants and their placementbecause many different load cases and alternatives can be simulated in a short time.

There are several possible improvements that can be made in the preliminary form ofthis research work. Firstly, approximated directional muscle forces can be replaced bythe real muscle model contributing passive stretching and active contraction. Secondly, asimple hyperelastic material model was implemented for the simulation of all PF tissues.More accurate results can be achieved with advanced fibre oriented anisotropic model.And finally, there has been some contradictory findings regarding the relative stiffness ofthe pelvic constituents. Besides an improved modeling additional experiments need to beconducted in order to derive reliable material parameters.

Acknowledgement

The authors have been partially funded by the Federal Min-istry of Education and Research through the FHprofUntproject BINGO. We would also like to thank our project part-ners: FEG Textiltechnik mbH and University hospital RWTHAachen for their regular suggestions and updated information.

REFERENCES

[1] von Hagens, G. Heidelberg plastination folder: collection of technical leaflets of plas-tination. Heidelberg, Germany, 2nd Edition, (1986).

[2] Petros, P.E.P. The female pelvic floor function, dysfunction and management accord-ing to the integral theory. Springer, 3rd Edition, (2010).

[3] Herschorn, S. Female pelvic floor anatomy: the pelvic floor, supporting structures,and pelvic organs. Rev. Urol. (2004) 6 (Suppl. 5):S2–S10.

[4] Petros, P.E.P. The pubourethral ligaments - an anatomical and histological study inthe live patient. Int. Urogynecol. J. (1998) 9 (3):154–157.

[5] Cruikshank, S.H. and Kovac, S.R. The functional anatomy of the urethra: role ofthe pubourethral ligaments. Am. J. Obstet. Gynecol. (1997) 176 (6):1200–1203.

[6] Zacharin, R.F. The suspensory mechanism of the female urethra. J. Anat. (1963) 97(3):423–427.

[7] Milley, P.S. and Nichols, D.H. The relationship between the pubo-urethral ligamentsand the urogenital diaphragm in the human female. Anat. Rec. (1971) 170 (3):281–283.

11


[8] DeLancey, J.O.L. Pubovesical ligament: a separate structure from the urethral sup-ports (“pubo-urethral ligaments”). Neurourol. Urodynam. (1989) 8 (1):53–61.

[9] Ingelman-Sundberg, A. The pubovesical ligament in stress continence. Acta. Obstet.Gyn. Scan. (1949) 28 (3-4):183–188.

[10] Williams, P.L., Warwick, R., Dyson, M. and Bannister, L.H. (Eds.). Gray’s Anatomy.Churchill Livingstone, UK, 37th Edition, (1989).

[11] Loubeyre, P., Copercini, M., Petignat, P. and Dubuisson, J.B. Levator ani musclecomplex: anatomic findings in nulliparous patients at thin-section MR imaging withdouble opacification. Radiology (2012) 262 (2):538–543.

[12] Azpiroz, F., Fernandez-Fraga, X., Merletti, R. and Enck, P. The puborectalis muscle.Neurogastroent. Motil. (2005) 17 (Suppl. 1):68–72.

[13] Rivaux, G., Rubod, C., Dedet, B., Brieu, M., Gabriel, B. and Cosson, M. Compar-ative analysis of pelvic ligaments: a biomechanics study. Int. Urogynecol. J. (2012)24 (1):135–139.

[14] Rubod, C., Brieu, M., Cosson, M., Rivaux, G., Clay, J.-C., de Landsheere, L. andGabriel, B. Biomechanical properties of human pelvic organs. J. Urology. (2012) 79(4):968.e17–968.e22.

[15] Rubod, C., Grosbras, P.L., Brieu, M., Giraudet, G., Betrouni, N. and Cosson, M. 3Dsimulation of pelvic system numerical simulation for a better understanding of thecontribution of the uterine ligaments. Int. Urogynecol. J. (2013) 24 (12):2093–2098.

[16] Chen, L., Ashton-Miller, J.A. and Delancey, J.O.L. A 3D finite element model ofanterior vaginal wall support to evaluate mechanisms underlying cystocele formation.J. Biomech. (2009) 42:1371–1377.

[17] Noakes, K.F., Pullan, A.J., Bissett, I.P. and Cheng, L.K. Subject specific finiteelasticity simulations of the pelvic floor. J. Biomech. (2008) 41 (14):3060–3065.

[18] Janda, S., van der Helm, F.C.T. and de Blok, S.B. Measuring morphological param-eters of the pelvic floor for finite element modelling purposes. J. Biomech. (2003) 36(6):749–757.

[19] Staat, M., Trenz, E., Lohmann, P., Frotscher, R., Klinge, U., Tabaza, R. andKirschner-Hermanns, R. New measurements to compare soft tissue anchoring sys-tems in pelvic floor surgery. J. Biomed. Mater. Res. B (2012) 4 (100):924–933.

[20] Frotscher, R. and Staat, M. Stresses produced by different textile mesh implants ina tissue equivalent. BioNanoMaterials, Special Issue Medical Textiles (2014).

12

Documents

A 3D FINITE ELEMENT MODEL OF THE FEMALE PELVIC FLOOR FOR THE RECONSTRUCTION OF URINARY ...congress.cimne.com/iacm-eccomas2014/admin/files/file... · 2014-03-14 · Key words: Urinary